Cause and affect

Michael Rutter looks at progress in the nature-nurture debate

The topic of nature and nurture continues to serve as a projective test eliciting responses that say as much about the person as about the scientific evidence. Nevertheless, the field is moving on in important and interesting ways. This disparate trio of books illustrates well some of the strengths, but also the complexity of the science and some of the accompanying hype and prejudice.

Lawrence Wright's well-written account of how twins can be used to study the different effects of genes and environment provides a good starting point. As expected from an award-winning science writer, the story is gripping, readable, easy to follow and scientifically well informed. It notes past abuses in the genetics field (and hence the need for continuing concerns on how genetic research is undertaken and used). Thus, there is brief mention of eugenic programmes, Josef Mengele's horrific experiments in Nazi Germany and Sir Cyril Burt's frauds in this country, but attention could have been drawn to the willingness of some of today's behavioural geneticists to have their research funded by organisations with a racist agenda.

The book goes on, however, to provide a positive appraisal of contemporary research. Quite rightly, attention is drawn to the consistent evidence that genetic influences account for a substantial proportion of the variation among individuals on most aspects of human behaviour - including social attitudes and personality characteristics. Indeed, genetic factors also influence how people shape, select and respond to their environments. As Wright argues, there is such a complicated amalgamation between genes and environments that any summary statement on the relative influence of each is necessarily a somewhat misleading oversimplification because it implies a separateness of their effects that does not match the reality. He rejects genetic and environmental evangelism but is forthright on the real value of understanding how genes influence susceptibilities and inclinations, how environments affect psychological development and functioning, and how the interplay between nature and nurture operates. The book ends with a brief but inevitably inconclusive discussion of what the findings mean for the concept of free will. Wright notes that the environment can be viewed as just as constraining of free will as genes. Neither, however, is determinative. There is still a degree of choice in how individuals direct their genetically and environmentally influenced tendencies toward better or worse uses.

The book accepts behavioural genetic claims that the common shared environment (such as the qualities of family life) has essentially no effect on the development of personality. Wright argues that this may be the most surprising and important discovery of the entire field of behavioural genetics. If true, that might be the case, but research during the past few years has forced a watering-down of the claim. As ordinarily measured, non-shared effects (meaning those that make children in the same family different from one another) include measurement error. Statistical techniques that take account of error have shown that non-shared effects are much less than had been claimed, especially with respect to traits as they are manifest over time. Also, shared environmental effects are much more important for some behaviours (for example, antisocial features) than for others (for example depression). Moreover, the findings regarding the strength of non-shared effects (even when true) do not mean that family influences are unimportant. Rather, what they mean is that experiences that seem to be family-wide (as would be the case with discord or neglect) impinge differentially on the children. When parents quarrel or fight, this may lead to one child, but not others, being scapegoated or drawn into the conflict.

Wright is also rather kind on the value of studies of separated twins. He suggests that they "stunningly tipped the balance in the nature-nurture debate". They did not; most geneticists are rather wary about relying on their findings because of the problems. The idea is splendid but the reality is that the samples are likely to be highly selective, the separation has often been only partial and the reporting leaves something to be desired.

What makes the book unusual and particularly interesting is the discussion of some of the many curiosities in twinning. Thus, Wright draws attention to the "vanishing twin" phenomenon by which one twin fails to develop, with the embryo being absorbed so that it disappears; to the possibility that prenatal influences may be more important than hitherto appreciated; to the association between twinning and congenital abnormalities; to the various factors (such as the age of the mother and the use of fertility drugs) that affect the rate of twinning; and to the finding that sometimes identical twins may not be genetically identical because they differ on which X chromosome is inactivated (so that only one twin may show the anomaly on the X chromosome). Wright notes the fascinating story of the identical twin boy who was raised as a girl because his penis was inadvertently destroyed by a circumcision error. The case was written up (and is much quoted) as showing the power of rearing on gender identity. More recently, however, it has been reported that during adolescence the "girl" rejected her female identity and insisted on surgery and reassignment as male. "She" is now "he" and is happily functioning as a married man with adopted children. In this case, nature exerted its power in the end. Of course, this is not a general message. The story does, however, constitute a warning about the dangers of jumping to premature conclusions.

This is not, nor does it attempt to be, an academic book. Nevertheless, Wright does a good job in highlighting some of the key scientific issues. Thus, he notes the under-representation of environmental high-risk environments in adoptee samples, and emphasises the difference between individual differences in liability (such as why John is delinquent and Peter is not) and variations in levels over time (such as the huge increase in crime of the past half century). Genetic influences are not likely to play much role in the latter despite their probable importance in the former. The book also underlines the fact that genetic findings do not support a determinist viewpoint and do indicate the probable role of complex indirect chain reactions. Thus it is completely implausible that there is a gene for crime but it may be that genetic factors play a role in the liability to anti-social behaviour by virtue of their effects on associated behaviours such as hyperactivity. Overall, the book provides an excellent introduction to some of the key issues and findings on twinning and a very good read into the bargain.

I started Ken Richardson's book with high hopes. As someone who has spent his professional lifetime studying psychological development, a developmental perspective on the nature-nurture debate was bound to be appealing, I thought. Regrettably, it is a very disappointing book. Two main problems dominate. First, it is hopelessly one-sided. Thus, Richardson argues that "there is little, if any, respectable scientific evidence that individuals or groups who vary in their cognitive abilities do so because of their genes". The evidence on group differences is inconclusive but that on individual variation is not. It is manifestly absurd to claim there is no evidence of genetic influences on variations in cognitive skills. Indeed, it is biologically ridiculous to suppose that there would be no such genetic influence. Richardson has some reasonable points on the limitations of twins and adoptee studies but his account is far from fair or balanced.

Second, there are multiple misrepresentations of scientific writing. For example, one of my papers is cited as strongly arguing for the value of data on separated twins. The paper referenced makes no mention of such data but a different paper, referenced by Richardson for a different point, makes explicit that I am actually very dubious about their value. Or, again, Richardson claims that intelligence is conceptualised as a fixed attribute; that would not apply to most writings on the topic. The account of how genes might influence normal traits (such as intelligence) also does not provide a true picture of the science.

All of this leaves one reluctant to accept other aspects of Richardson's claims. That is a pity because, although put in a one-sided fashion, his book raises some most important considerations. I am least impressed by his critique of the fact that intelligence is an inferred construct, the exact nature of which is difficult to pin down. That is correct but it applies to most psychological constructs and the query has to be whether the inference seems valid and useful. Richardson is on much stronger ground when urging the importance of understanding that many genetic factors influence processes of development rather than fixing the reaction range of traits. Many geneticists make the same point but Richardson is rather vague in his account of what might be involved. He is on target with his point that a high heritability in no way implies that environmental changes cannot make a big impact, and he is rightly scathing about some of the genetics hype - such as the ludicrous claim that the isolation of the first gene involved in determining intelligence will be a turning point in human history. It will not, because it is likely that any single gene will have only a tiny effect and, hence, may shed little light on the brain processes involved, and because it cannot be assumed that the gene is directly influencing some neural mechanism (such as the speed of information processing) rather than some attribute (such as a specific cognitive component or a temperamental feature) that is more indirectly implicated. There is as yet no replicated finding of a susceptibility gene for intelligence but, when there is, it will constitute a tiny useful step in the right direction rather than a turning point in history. There is, as Richardson notes, a misleading mystique about intelligence but his case is greatly overstated.

Eleanor Maccoby's wonderful book on gender differences is quite different from the other two in that it includes very little direct discussion of genetics. In fact, it is unfortunate that she is completely wrong about the only thing she does say, namely, that twin studies cannot be used to study gender differences because they include only same-sex pairs. On the contrary, geneticists regard opposite-sex pairs as especially useful just because they can shed light on gender differences with respect to traits present in both sexes but which differ in level or pattern between males and females.

What Maccoby does supremely well, however, is discuss how a biologically influenced propensity on which boys and girls differ to a minor degree can nevertheless lead to quite large differences in behaviour through their indirect influence brought about by effects on interaction styles and social groups. She skilfully integrates qualitative and quantitative research findings, with helpful verbatim quotes on how children talk about themselves and their social interchanges, and pursues a carefully argued discussion of what the research might mean. The academic critique is of the highest order but the style is flowing, engaging and intensely interesting. The book is provocative in forcing a rethink on gender differences and challenging in its conclusions.

The basic message is that biology plays an important contributory role in gender differences, but the differences between boys and girls are quite small for the most part. On the whole, child-rearing influences are of very minor importance with respect to these small differences in behavioural traits. Maccoby argues that there are three major phenomena that need to be explained:

* gender segregation (the strong tendency from three years on for children to play largely with others of their own sex);

* differentiation of interaction styles (such that the boys' groups and girls' groups differ very greatly in how they function);

* asymmetry (that boys' groups tend to be more cohesive and exclusionary than girls' groups). Maccoby is persuasive in her claim that gender distinctions arise mainly in social interactions and that peer groups are highly influential in greatly enhancing gender differences in behaviours such as aggression and dominance. When boys and girls are by themselves their behaviour contrasts only slightly; males behave differently when interacting with a female than when with another male; and all-male groups differ radically from all-female groups in their behaviour. Maccoby's book provides a fascinating account of how this comes about during the course of development and how it operates during adult life as well as childhood. This is a really important book for anyone concerned to understand psychological development. The findings and concepts have crucially important implications, too - although they are not addressed directly- for the understanding of gender differences in problem behaviours such as crime and depression. Similarly, there are implicit messages for how development is involved in nature-nurture interplay.

Sir Michael Rutter is professor of developmental psychopathology, Kings College, London.